The short answer: that’s your firm’s decision. What alfred_ can give you are the facts your compliance team will ask for. It sends email only after you approve it, from your own mailbox, so sent mail is in your Sent folder like any other. It isn’t an archiving or supervision system. Your texts with it are with the assistant, not with clients. And it doesn’t use your data to train AI models.
Why advisors ask first
You’re an advisor at an independent RIA with four advisors and two client service associates. Half your day is client email: confirming review meetings, answering “did the transfer go through?”, sending the paperwork for a beneficiary change. An assistant that tells you which client emails need you and writes the replies would give you hours back.
Then you remember the compliance manual. Client communications are supervised and archived. Unapproved channels are a problem. A tool that might send something on its own, or keep client data where your archive can’t see it, is a conversation with your chief compliance officer, not a download.
What alfred_ does with client email
It reads the inboxes you connect. You sign in with Google or Microsoft, so your password stays with them and alfred_ doesn’t receive it, and you can remove its access from that account at any time.
It sends only what you approve, from your mailbox. When you ask for a reply, alfred_ writes it in the client’s thread and shows it to you. It goes only after you approve that email, and it goes out through your own Gmail or Microsoft 365 account, so it’s in your Sent folder like any email you wrote yourself.
Nothing sends on its own, unless you set up a rule that forwards email or a routine you tell to send without asking. If your firm requires pre-approval, don’t set one up, and every client email waits for you.
It doesn’t train on your data. alfred_’s privacy policy says it does not use your data to train or fine-tune AI models.
What alfred_ is not
- Not an archive. alfred_ doesn’t replace your archiving or e-communications supervision. Because it sends through your mailbox, whatever captures your mailbox today sees that mail the same way. Confirm that with compliance.
- Not a client texting channel. Clients don’t get texts from alfred_. When you text alfred_, you’re instructing your assistant, and the result is an email you approve.
- Not connected to your custodian, CRM of record or planning software unless you connect a supported CRM (HubSpot or Salesforce), and then it changes only what you approve.
Questions to take to your compliance team
- Does our policy allow a third-party app with read access to firm email, signed in through Google or Microsoft?
- Does our archive capture email sent through Gmail’s or Microsoft’s own sending (how alfred_ sends), as it does mail sent from Outlook or Gmail directly?
- Do texts between an advisor and an assistant (not with clients) fall under our text-messaging policy?
- Do we require pre-approval of certain client communications, and is approving each email in alfred_ enough for that?
If your firm’s Microsoft 365 or Google Workspace requires IT approval for new apps, your admin can approve alfred_ once from its IT admin consent page.
The honest caveat
Nothing here is legal or compliance advice, and alfred_ can’t certify that it meets any regulator’s rules for your firm. It can tell you how it works: approval before sending, sending from your mailbox, no training on your data, and no archive of its own. Read the privacy policy and security page before connecting a firm inbox. alfred_ is a paid product; plans are on the pricing page.
